LLaDA
Benchmarks
TaskDatasetSOTA resultTrendResults
LLaDA-MoE
21.2TPS (tokens/sec)
7
LLaDA mini 16B (32K context) 2.0
2,717Throughput (tokens/s)
4
LLaDA 16B 16K context 2.0-mini
3,475Throughput (tokens/s)
4
LLaDA mini 16B 8K context 2.0
3,813Throughput (tokens/s)
4